Populist PortalMicrosoft's portal debut is document management oriented, departmentalBy Colin White Continued from Page 1 Every document has a document profile of searchable metadata. When a document is checked in, SharePoint requests the user to complete the profile, which can contain both mandatory and optional fields. By default, this profile includes basic properties such as author and title, but organizations add custom properties to the profile to capture additional information and thus make it easier to locate. The big advantage of SharePoint profiles over the current unused Microsoft Office properties facility is that metadata entry can be enforced: An author can't save a document before completing the mandatory fields. Roles are used to control access to documents. Coordinators perform management tasks, authors add and update documents, and readers have read-only access to published documents. SharePoint also has the option of denying a user access to specific documents. Standard Windows security facilities are used to authenticate SharePoint users. ACCESSING BUSINESS CONTENTDocuments in SharePoint workspaces can be accessed using Microsoft Office, Windows File Manager, or Internet Explorer. All three interfaces provide the ability to navigate and search a workspace's folder structure. This review is restricted to a discussion of accessing documents using a Web browser and the dashboard site, since, from a portal perspective, this will be the normal method of accessing a combination of SharePoint and non-SharePoint business content. I already explained how you use Web parts to access business content. A URL Web part can be used, for example, to point to a specific information service. Similarly, a PowerPoint presentation in the SharePoint workspace could be saved as a Web part for display through the dashboard site. To help users locate business content, Microsoft also provides predefined category, search, and subscription Web parts (see Figure 1). Business content viewed via the dashboard site can be organized by business category. Users who are unfamiliar with the business content can use the category Web part to browse portal content by business topic. The category facility also lets any given piece of content be associated with multiple business topics. SharePoint categories can be associated with documents maintained in SharePoint workspaces and by non-SharePoint content servers. You can assign a document to a category manually or let the SharePoint Category Assistant do it automatically. After the user has categorized a few representative documents for each category, the Category Assistant is trained against these documents, so that algorithms can be established for automatically categorizing new content. It remains to be seen how usable and effective this feature is. The search Web part lets the portal user enter a query to search any business content that has been indexed by SharePoint search service. Both keyword and full-text searches are supported. Results returned from the search service are ranked using a probabilistic ranking algorithm. There are four components to the SharePoint search service: gather, filter, index, and search. The gather component automatically collects content and metadata from SharePoint workspaces, Web servers, file servers, Exchange public folders, and Lotus Notes databases. A supplied development kit can be used to develop "protocol handlers" for other content sources. As supplied, the gather component doesn't support OLE DB or ODBC. An Active Server Page script could be used, however, to access database data. The gather process can be initiated manually or on a scheduled basis. The filter component is used to decode specific file formats in the gathered content. Filters are provided for HTML, office documents, text, and fax documents in TIFF format. Adobe also provides a filter for PDF files. Custom filters can be developed using the supplied development kit. The index component creates a searchable index by using language-specific word breakers and "stemmers" to extract words from the gathered content. (A stemmer takes a word, breaks it down to its root by removing all suffixes, and searches for similar words based on the root word.) The search component uses SQL full-text extensions to access the SharePoint index, which is stored in the Web Storage System. Several search APIs - ADO, OLE DB, and WebDAV - are provided to allow programmatic access to the index. Another way to access business content is to subscribe to it. Subscriptions notify users about new or updated business content. A subscription may be associated with a specific document, all the documents in a folder, all documents assigned to a category, or a set of search results. Users can view subscription notifications on the dashboard site, or they can receive notifications by email. A DOCUMENT MANAGEMENT CONTENDERThe first release of the Microsoft SharePoint Portal Server is a significant addition to the portal marketplace, and the product is likely to have a major impact on competitive products that offer portals focused toward information delivery and document management. Unlike IBM and Oracle, which have integrated portal technology into their Web application servers, Microsoft has decided to offer a separate portal product that has its own document management system and is tightly integrated with Microsoft Office XP. This strategy is clearly designed to be competitive with offerings from Lotus for Lotus Notes users. Colin White [cwhite@databaseassociates.com] is president of DataBase Associates International and conference chairperson of the DCI Corporate and E-Business Portals conference.
|
 |
Most Popular This Week
IE Weekly Newsletter
Subscribe to the newsletter
| |
| |||||||||||||||||||||||||||||||
